Exploring the integration of artificial intelligence in radiology education: A scoping review.

BACKGROUND: The integration of Artificial Intelligence (AI) into radiology education presents a transformative opportunity to enhance learning and practice within the field. This scoping review aims to systematically explore and map the current landscape of AI integration in radiology education. METHODS: The review process involved systematically searching four databases, including MEDLINE (Ovid), Embase (Ovid), PsychINFO (Ovid), and Scopus. Inclusion criteria focused on research that addresses the use of AI technologies in radiology education, including but not limited to, AI-assisted learning platforms, simulation tools, and automated assessment systems. This scoping review was registered on Open Science Framework using the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) extension to scoping review. RESULTS: Of the 1081 search results, 9 studies met the inclusion criteria. Key findings indicate a diverse range of AI applications in radiology education, from personalized curriculum generation and diagnostic support tools to automated evaluation systems. The review highlights both the potential benefits, such as enhanced diagnostic accuracy, and the challenges, including technical limitations. CONCLUSION: The integration of AI into radiology education, which has significant potential to enhance outcomes and professional practice, requires overcoming existing challenges and ensuring that AI complements rather than replaces traditional methods, with future research needed on longitudinal studies to evaluate its long-term impact.

ChatGPT and assistive AI in structured radiology reporting: A systematic review.

INTRODUCTION: The rise of transformer-based large language models (LLMs), such as ChatGPT, has captured global attention with recent advancements in artificial intelligence (AI). ChatGPT demonstrates growing potential in structured radiology reporting-a field where AI has traditionally focused on image analysis. METHODS: A comprehensive search of MEDLINE and Embase was conducted from inception through May 2024, and primary studies discussing ChatGPT's role in structured radiology reporting were selected based on their content. RESULTS: Of the 268 articles screened, eight were ultimately included in this review. These articles explored various applications of ChatGPT, such as generating structured reports from unstructured reports, extracting data from free text, generating impressions from radiology findings and creating structured reports from imaging data. All studies demonstrated optimism regarding ChatGPT's potential to aid radiologists, though common critiques included data privacy concerns, reliability, medical errors, and lack of medical-specific training. CONCLUSION: ChatGPT and assistive AI have significant potential to transform radiology reporting, enhancing accuracy and standardization while optimizing healthcare resources. Future developments may involve integrating dynamic few-shot prompting, ChatGPT, and Retrieval Augmented Generation (RAG) into diagnostic workflows. Continued research, development, and ethical oversight are crucial to fully realize AI's potential in radiology.

Radiologist preferences for faculty development initiatives to improve resident feedback in the era of competency-based medical education.

INTRODUCTION: Since 2022, all Canadian post-graduate medical programs have transitioned to a Competence by Design (CBD) model within a Competency-Based Medical Education (CBME) framework. The CBME model emphasized more frequent, formative assessment of residents to evaluate their progress towards predefined competencies in comparison to traditional medical education models. Faculty members therefore have increased responsibility for providing assessments to residents on a more regular basis, which has associated challenges. Our study explores faculty assessment behaviours within the CBD framework and assesses their openness to opportunities aimed at improving the quality of written feedback. Specifically, we explore faculty's receptiveness to routine metric performance reports that offer comprehensive feedback on their assessment patterns. METHODS: Online surveys were distributed to all 28 radiology faculty at Queen's University. Data were collected on demographics, feedback practices, motivations for improving the teacher-learner feedback exchange, and openness to metric performance reports and quality improvement measures. Following descriptive statistics, unpaired t-tests and one-way analysis of variance were conducted to compare groups based on experience and subspecialty. RESULTS: The response rate was 89% (25/28 faculty). 56% of faculty were likely to complete evaluations after working with a resident. Regarding the degree to which faculty felt written feedback is important, 62% found it at least moderately important. A majority (67%) believed that performance reports could influence their evaluation approach, with volume of written feedback being the most likely to change. Faculty expressed interest in feedback-focused development opportunities (67%), favouring Grand Rounds and workshops. CONCLUSION: Assessment of preceptor perceptions reveals that faculty recognize the importance of offering high-quality written feedback to learners. Faculty openness to quality improvement interventions for curricular reform relies on having sufficient time, knowledge, and skills for effective assessments. This suggests that integrating routine performance metrics into faculty assessments could serve as a catalyst for enhancing future feedback quality.

Incidence of left ventricular thrombus following STEMI in the modern era via multimodality imaging: A systematic review and meta-analysis.

Background: Left ventricular thrombus (LVT) is a significant complication in STEMI. Previous studies were conducted prior to modern timely percutaneous reperfusion networks. Current expert opinion suggests incidence in the current era has decreased. We conducted a systematic review and meta-analysis to better understand the incidence and diagnosis of LVT in patients with STEMI treated with timely percutaneous techniques as assessed by multimodality imaging. Methods: Cochrane, EMBASE, LILACS, and MEDLINE were searched over the last 10 years only including studies using contemporary techniques. The primary outcome was detection of LVT in patients via echocardiogram with or without contrast or Cardiac MRI (cMRI) following STEMI (both anterior and any territory) treated with PCI. Data was pooled across studies and statistical analysis was conducted via random effects model. Results: 31 studies were included. 18 studies included data on any territory STEMI, totaling 14,172 patients, and an incidence of 5.6% [95% CI 4.3-7.0]. 18 studies were included in analysis for anterior STEMI, totaling 7382 patients and incidence of 12.7% [95% CI 9.8-15.6]. Relative to cMRI as a gold standard, the sensitivity of non-contrast echocardiography to detect LVT was 58.2% [95% CI 46.6-69.2] with a specificity of 97.8% [95% CI 96.3-98.8]. Conclusions: Incidence of LVT in STEMI patients treated with contemporary timely percutaneous revascularization is in keeping with historical data and remains significant, suggesting this remains an ongoing issue for further investigation. Numerically, both cMRI and contrast echo detected more LVT compared to non-contrast echo in any-territory STEMI patients.

Prophylactic Anticoagulation to Prevent Left Ventricular Thrombus Following Acute Myocardial Infarction: A Systematic Review and Meta-Analysis.

Clinical practice guidelines from the American Heart Association recommend consideration of prophylactic anticoagulation to prevent left ventricular thrombus (LVT) formation in patients with anterior ST-elevation myocardial infarction. These guidelines were given a low certainty of evidence (class IIb, level C), relying primarily on case studies and expert consensus to inform practice. Our objective was to compare the safety and efficacy of prophylactic anticoagulation, in addition to dual antiplatelet therapy, in the current era of timely primary percutaneous coronary intervention. Electronic databases, including EMBASE, MEDLINE, and Cochrane Library, were systematically searched from January 2012 through June 2022. A total of 7,378 publications were screened, and 5 publications were eventually included in this review: 1 randomized control trial and 4 retrospective studies involving 1,461 patients. Data were pooled using a fixed-effects model and reported as odds ratios (ORs) with 95% confidence intervals (CIs). The primary outcome of interest was the rate of LVT formation, and the secondary outcomes were the rate of major bleeding and systemic embolism. Pooled analysis showed a significantly lower rate of LVT formation (OR 0.28, 95% CI 0.11 to 0.73, p <0.01) and significantly higher rates of bleeding (OR 2.85, 95% CI 1.13 to 7.24, p = 0.03) in the triple therapy group compared with dual antiplatelet therapy. No significant difference was observed in the rate of systemic embolism between the groups (OR 0.37, 95% CI 0.12 to 1.13, p = 0.08). In this meta-analysis, there is no conclusive evidence to either support or oppose the use of triple therapy for LVT prevention in patients with anterior ST-elevation myocardial infarction treated with primary percutaneous coronary intervention. Appropriately powered randomized controlled trials are warranted to further evaluate the benefits of LVT prevention against the risks of major bleeding in this population.